With the advent of recombinant DNA technology and cell culture techniques in recent years, the controlled biological production of useful and pharmacologically interesting compounds, especially proteins, such as interferon, insulin and antigens, has become possible. There is an increasing need for the development of biological systems ensuring the large scale production of further proteins of biological, especially pharmacological, interest.
The term "protein" as used hereinbefore and hereinafter is intended to include polypeptides of high molecular weight, e.g. of over about 34000, and also polypeptides of lower molecular weight, e.g. of below about 34000, and derivatives thereof, such as glycosylated, phosphated and sulfated derivatives.
The advances in recombinant DNA technology make it feasible to introduce the gene encoding a desired protein into microorganisms and then induce the microorganisms to synthesize the protein. However, many biologically important molecules cannot be synthesized by this technology. This is especially true for those molecules the structure of which is not yet known. In most cases, the proteins secreted by genetically modified microorganisms are not a faithful replica of the authentic molecules but differ from the latter with respect to the N and C termini of the amino acid sequence. This fact is due to the experimental procedure in recombinant DNA technology. Furthermore, glycosylated proteins cannot be produced by microorganisms, such as bacteria and to a certain extent also yeast, which lack the necessary cellular machinery. In many cases, cell and tissue culture techniques can be advantageously made use of. As cell cultures originate from intact organisms the proteins produced by the cell cultures correspond to the naturally occuring proteins in all respects.
However, cultivation of cells of higher organisms, such as mammalian cells, on a large scale is a difficult problem. The nutrient requirements of such cells are more stringent than those of most microorganisms which proliferate in artificial media. The growth medium of most mammalian cells described so far has to include serum which is very expensive. The cost of serum largely determines the economic feasibility of the cell culture technique and may limit its applicability to the production of proteins which are not available otherwise. Some cells can be cultured in serum-free medium supplemented with hormones or growth factors such as transferrin, insulin, epidermal, fibroblast or nerve growth factor. In most cases, however, cells will not multiply indefinitely in these serum-free media.
Cell lines proliferating in a serum-free medium are of particular importance for the production of such proteins which are susceptible to destruction or contamination by serum or exogenously added growth factors. Such a protein is for example pro-tissue plasminogen activator (pro-TPA).
The so-called plasminogen activators, have become the subject of scientific investigations showing their evident clinical applicability in the lysis of blood clots. Blood clots are composed of fibrin which has been formed from its soluble precursor, fibrinogen, under the action of the enzyme thrombin. They are one of the major causes of morbidity and of mortality in humans and dissolving them without side effects is difficult to achieve.
Mammalian plasma contains an enzymatic system capable of dissolving the fibrin in blood clots. One component of the fibrinolytic system consists of the enzymes, plasminogen activators, which convert plasminogen (an inactive proenzyme form of plasmin) into the proteolytic enzyme plasmin. Plasmin then degrades the fibrin network of the clots to form soluble products. In cases where the thrombolytic potential of the body is insufficient to remove intravascular thrombi formed, for example in patients suffering from thromboembolisms or post-surgical complications, it may be indispensable to use exogenously administered thrombolytic agents.
There are two activators of human plasminogen which are commercially available for thrombolytic therapy: urokinase, a serine protease isolated from human urine or cultured kidney cells, and streptokinase, a bacterial protein obtainable from streptococci. Since neither enzyme has a specific affinity for fibrin, thrombolysis with these substances is associated with systemic activation of plasminogen which can produce indiscriminate digestion of coagulation proteins, and significantly increase the risk of internal bleeding (hemorrhage) during treatment. Another disadvantage of urokinase is its very short useful half-life following its injection into humans. For this reason, high doses of urokinase are needed to achieve effective fibrinolysis. Streptokinase being a protein foreign to man, gives rise to the production of neutralizing antibodies which block its action, and to allergic reactions which are harmful and potentially fatal.
Another group of plasminogen activators, called tissue plasminogen activators (hereinafter referred to as "TPAs") are known to exist in most human tissues. TPAs originating from different tissues possibly differ from each other with respect to their molecular properties but are immunologically similar. They differ from urokinase with respect to their chemical and immunological properties, in their greatly enhanced fibrinolytic action in the presence of fibrin, as well as in their high affinity for fibrin [cf. S. Thorsen et al., Thronb. Diath. Haemorrh. 28, 65-74 (1972), D. C. Rijken and D. Collen, J. Biol. Chem. 256, 7035-7041 (1981)]. Because of their high affinity for fibrin, the action of TPAs is confined to the locality of the clot thereby reducing significantly the danger of uncontrolled hemorrhage.
The following scheme shows the relationship between plasminogen, Plasmin, fibrin and the various plasminogen activators. ##STR1## Recently, two patients suffering from coagulation disorders have been successfully treated with a PA isolated from the culture fluid of a human melanoma cell line [cf. W. Weimar et al., The Lancet (1981). 1018-1020]. There are two molecular forms of TPA: the active twochain form and an inactive one-chain form [precursor TPA or "pro-TPA"; for reference, cf. D. C. Rijken and D. Collen, loc. cit.; D. C. Rijken et al., J. Biol. Chem. 257, 2920-2925 (1982) and P. Wallen et al. Progr. Fibrin. 5, 16-23 (1982)]. Pro-TPA can be converted to active TPA by incubation with fibrin or by the influence of plasmin which by this cascade-like reaction triggers its own synthesis.
Sources of human TPA include extracts of various human tissues (which are not available for commercial exploitation) and various human tumor cells which have been shown to release TPAs to a varying extent [E. L. Wilson et al., Cancer Research 40, 933-938 (1980); E. L. Wilson et al., Blood 61, 568-574 (1983)].
In a recently filed patent application (EP 41766, inventors D. Collen, D. C. Rijken and O. Matsuo) a TPA with a molecular weight of 72000 is disclosed which has been isolated from a cultured human melanoma cell line (Bowes) and which is probably identical to the TPA already described by E. L. Wilson et al. [Cancer Research 40, 933-938 (1980)]. Like other cell lines known so far, this human melanoma cell line, Bowes, requires the presence of serumfor growth, e.g. foetal calf serum. However, serum is very expensive and contains proteinaceous components which contaminate the produced TPA and prevent the isolation of high amounts of pro-TPA. This may lead to a tedious and laborious purification procedure for either TPA or pro-TPA.